Killing Me Softly

Just over a year ago, we wrote about a behavioural bias called the “freak-out” effect where sharp market downturns cause investors to “panic sell.” The authors of the paper studying the freak-out effect found that of the investors who engaged in panic-selling behaviour, 30.9% of these investors never reinvested in risk assets (i.e., the stock market). The research also suggested that for investors who panic sold, only half would re-enter the market within six months. Other than the troubling statistic that panic selling causes some investors to never reinvest in the stock market and therefore need to deal with inflation eroding the purchasing power of cash savings, the act of selling risk assets during a drawdown is neither inherently good nor bad. The result of the decision is only observable after the fact and, in this example, the result is highly dependent on what type of market structure is prevalent. For example, the same decision to panic sell during the Global Financial Crisis would have resulted in a wildly different outcome (positive) than panic selling during the COVID-19 drawdown (negative). I stated at the time that the natural extension of the work on panic selling would be to look at how utilizing a rules-based approach could guard against behavioural biases and help protect against the negative consequences of the freak-out bias.

The reason is that the act of panic selling is only half of the equation. I would further posit that one of the reasons that such a large percentage of investors never return to risk assets after panic selling is a lack of a framework for how to re-enter the market. For example, if you panic sold at the bottom of the COVID-19 drawdown in late March 2020, what is your decision-making criteria to identify how and when to re-enter the market? After the initial decision to panic sell, you are now faced with another behavioural bias that wreaks havoc on decision making – the disposition effect. The disposition effect shows that individuals dislike losses more than they enjoy gains. In financial markets, the disposition effect is generally explained as investors holding on to their “loser” stocks for too long, using the positive narrative they have concocted to justify continue holding the position. How many times have you heard, “if I liked this stock at $50, then I really love it at $40”? The disposition effect is not just a behavioural bias that affects retail investors; this Wall Street Journal article explores a recent academic paper that shows that professional money managers lose an average of 0.8% annually through sub-optimal selling decisions driven by either cutting winners too early or holding on to losers too long.

Even though academics tend to describe the disposition effect as holding on to a bad stock pick for too long, the same mental accounting can be applied to any trading decision, especially one where your reference point gets flipped to being “long cash.” In the example where an investor has panic sold during the COVID-19 drawdown, their reference asset now becomes their new portfolio, which is long cash and short stocks (i.e., the investor sold stocks and received cash). If stocks start to rally right after you panic sold, what is your decision-making criteria to now tackle the disposition effect of when to sell your cash and buy stocks to re-enter the market? How are you determining that your initial decision to sell was wrong (as an aside, there is nothing wrong with embracing the concept of error in order to eliminate future mistakes) and what data are you using to improve your process. Annie Duke introduces the topic of “kill criteria” in her new book Quit: The Power of Knowing When to Walk Away, which is essentially another way of designing rules-based criteria in advance of a decision in order to steer your future decision making in the heat of the moment.

At Viewpoint, we wholeheartedly agree with Annie Duke, as we purposefully design rules-based criteria to eliminate the impact of behavioural biases in the heat of the moment. We also emphasize the importance of being disciplined in research, reaching conclusions, and making decisions in “cold blood” before codifying specific rules to respond to various market conditions in a rational way. These factors embody the Viewpoint principle that “disciplined, rules-based strategies are ideal for combatting decision-making biases.” These behavioural biases plague both retail and professional investors, and the research shows that only being aware of behavioural biases is insufficient to mitigate those biases. Instead, putting rules in place to alter the “choice architecture” is more likely to improve rational decision making.